idea-010 · narrative
A synthesis across this candidate's claims, dossier and decks — every statement here traces to something that already exists elsewhere in the knowledge base. A narrative never invents a fact, upgrades a tag, or recommends anything.
How to read this. This is a synthesis, not a new source of truth: every claim it cites carries the tag the candidate file gives it right now, copied not paraphrased, and every line quoted from a deck is quoted verbatim. See docs/narrative-spec.md.
What this build read, and what it did not. The candidate file in full (68 claims), the one dossier its domains: lists (potassium-monitoring-ckd, last refreshed 2026-08-26), and all three decks (each built 2026-09-29; check_decks.py reports all three current). The candidate lists no macro-trend seeds, so no seed context appears below. Research reports were not read directly; they reach this page only where a line has become a claim.
This is a single-use strip and handheld meter that reads potassium and creatinine from one fingerstick capillary drop. A third on-strip channel exists only to detect a hemolysed sample and suppress the potassium result rather than report a falsely high one. It is meant for trained staff in outpatient heart-failure and CKD (chronic kidney disease) clinics, during RAAS-inhibitor dose titration. Since the last build, the Verifier has moved three claims to verified on 2026-09-29. FDA (US Food and Drug Administration) operates a Dual 510(k) (premarket notification) and CLIA (Clinical Laboratory Improvement Amendments) Waiver by Application pathway claim 8verified. The two lab codes are paid from the Clinical Laboratory Fee Schedule, at $4.76 and $5.12 in 2026 claim 14verified. And Medicare fee-for-service volume for the two standalone codes is on file as a floor claim 16verified. That gives the file its first revenue-side numbers, but no cost, share, margin or calendar figure is verified. The question that decides whether the product can exist has not moved: whether a fingerstick potassium can be screened for hemolysis without voiding too many results to be usable claim 22unconfirmed, claim 23unconfirmed. All three rebuilt decks now converge on that suppressed-result rate from different directions.
The problem. RAAS inhibitors and MRAs are titrated against potassium. The file's clinical premise is that fear of hyperkalemia leads clinicians to underdose, stop and fail to restart these drugs. The cessation figures behind that premise came from a search summary, and no primary paper was read claim 20unconfirmed. The monitoring cadence that would make an in-visit result recur (baseline, about 2 weeks, about 1 month, then periodically) comes from a guideline summary, not from the guideline itself claim 21unconfirmed. The specific moment the device targets, where the prescriber lacks a same-visit potassium result and defers the dose decision, has not been checked by anyone claim 60unverified. The dossier holds context on the drug side that has not become a claim. Several cohorts record RAASi discontinuation or down-titration after hyperkalemia, but every study that isolates hyperkalemia as the stated reason is non-US. The one US registry located (CHAMP-HF) reports general discontinuation, with CKD as the only independent predictor across drug classes [dossier: potassium-monitoring-ckd].
Why now. No seed is linked to this candidate, and nothing in the file or the dossier evidences a change in the world that makes this newly possible. What has changed is the pipeline's own reach. CMS (Centers for Medicare & Medicaid Services) payment, utilization and coverage caches and a De Novo (De Novo classification request) listing now exist locally, and that is why claims 14 and 16 moved on 2026-09-29. The file says so itself: "our reach improved, the world did not change".
The mechanism. Capillary fingerstick sampling lyses red cells, hemolysis releases intracellular potassium, and the resulting false elevation is the file's named reason that no strip-based fingerstick potassium meter already exists. That is the Generator's reasoning, with no source read and no connector able to reach it claim 22unconfirmed. The dossier's literature rows describe the problem without testing any product. On laboratory and blood-gas analysers, capillary samples hemolysed at 17.7% against 10.6% for venous, and potassium was one of the analytes that failed capillary-versus-venous interchangeability [dossier: potassium-monitoring-ckd]. The design answers with suppression, not correction, so it trades a wrong answer for no answer. The file names the real performance metric: the invalid-result rate in ordinary clinic hands, where a device voiding one draw in four is not usable claim 23unconfirmed. Whether an on-strip hemoglobin channel can detect free hemoglobin at the concentration that biases potassium is bench physics claim 40unconfirmed. So is whether a plasma-equivalent potassium calibration holds across the hematocrit range of HF and CKD patients claim 41unconfirmed.
The regulatory shape. The skeleton is verified. Potassium is 21 CFR (Title 21 of the Code of Federal Regulations) 862.1600, Class II (FDA device class II), whose identification names serum, plasma and urine but not capillary whole blood claim 1verified. Creatinine is 21 CFR 862.1225, Class II, naming only plasma and urine claim 2verified. The product codes are CEM claim 3verified and CGL claim 4verified. The creatinine channel has a cleared strip-and-meter predicate, K070068 (an FDA 510(k) clearance number). Its IFU (instructions for use) covers capillary whole blood at point of care and is calibrated to plasma-equivalent results, which is how a whole-blood device has sat inside a plasma-worded generic type claim 5verified. A second Nova fingertip strip-and-meter, a glucose/β-ketone device cleared for home use, is a format analog only claim 6verified. The dual pathway exists; the claim settles the route and not the study package it requires claim 8verified. The potassium channel is the unresolved half. The audit enumerated both product codes under 862.1600 over thirty years and found no strip-and-fingerstick device. It also found cleared whole-blood point-of-care potassium systems (K200865, K042270), so the frontmatter's "no specific predicate identified for the potassium channel" is recorded as overstated against the candidate's own interest claim 7unconfirmed. The file's premise-correcting claim follows the same line. A cleared Piccolo potassium system exists, its CLIA-waived status was never checked, and the gap narrows to form factor and per-test cost claim 10unconfirmed.
The money. The Clinical Laboratory Fee Schedule amounts are now known for Medicare. The cost of producing a reported result in a clinic is not, so the per-test inequality is unresolved claim 36unconfirmed. The CPT (Current Procedural Terminology) descriptor wording that would let existing codes fit a capillary strip is AMA-licensed and unchecked claim 11unconfirmed, claim 12unconfirmed. The business case and the viability deck have added questions the Verifier has not yet seen. Are the target clinics physician-office labs or hospital-owned departments billing on the hospital's claim claim 57unverified? Do non-Medicare payers match CLFS (clinical laboratory fee schedule) claim 58unverified? Can a suppressed result be billed at all claim 63unverified?
What has and hasn't been checked. Everything verified is a regulatory-database fact or a Medicare payment or volume fact. Nothing verified speaks to whether the potassium channel works, whether anyone would change a dose on it, or what it costs to run. The candidate's own Verified Composite and Gate Check are dated 2026-08-25 and were computed when 6 claims were verified. Nine are verified now, so those sections are behind the tags they sit beside. This narrative neither quotes nor re-derives them. The candidate's Wrap section is also behind: it counts "6 of 43 claims verified" and quotes the 2026-08-25 decks' load-bearing conditions, which the 2026-09-29 decks have since replaced. Fixing either is a stage's job, not this page's.
Only claims the candidate file tags verified.
No claim in the file is refuted. Everything below is unconfirmed (the Verifier attempted it and could not settle it) or unverified (appended on or before 2026-09-29 and never checked).
Mechanism and bench physics: the factor the file says decides the candidate.
Regulatory: the potassium half.
Reimbursement and who gets paid.
data/ as holding no CMS files. That is behind the candidate file, whose claims 14 and 16
now read those caches [dossier: potassium-monitoring-ckd]
Money, market and calendar.
People: prescribers, staff, patients, purchasers.
Clinical premise, IP (intellectual property) and field.
PATENTSVIEW_API_KEY is unset claim 25unconfirmed. No in-force US patent claims hemolysis
detection with co-located potassium suppression claim 68unverified
All three decks were rebuilt on 2026-09-29 against the candidate as it stands now, and check_decks.py reports each one current. None is missing and none is STALE. They agree on one structural fact: every verified claim is the known half of a condition and never the half it turns on. They also converge on one number from three sides. The feasibility deck's suppressed-result rate becomes the desirability deck's repeat-stick burden and the viability deck's unbillable-strip cost. They disagree on what can be checked before a strip exists. The desirability and viability decks each say their own condition can be asked first. The feasibility deck says its condition can only be answered by bench and multi-operator data.
Desirability. The chair is three users at once: prescriber, operator and patient. Its new claims put the "deferred decision" moment on record claim 60unverified, along with laboratory-director approval claim 61unverified and guideline acceptance claim 62unverified. It is explicit that the three moved tags say nothing about whether anyone wants the device. Every stakeholder claim it rests on is unconfirmed or unverified, because adoption returns unverifiable by design. Its one proposed pre-device proxy is retention of incumbent point-of-care analyzers at twelve months claim 59unverified. It names a single condition that every other slide depends on, and argues that condition can be asked before any strip exists. The same claim, cited fresh with its current tag, is claim 28unconfirmed.
"would an outpatient HF or nephrology prescriber change a RAASi or MRA dose within the visit on a fingerstick potassium result, or order a confirmatory venous draw and defer the decision anyway" — the load-bearing condition of the desirability deck (built 2026-09-29).
Viability. This is the lens the tag moves touched most. Revenue per code is now verified claim 14verified and the Medicare volume floor is on file claim 16verified. The deck says plainly: "The cost side is blank." It adds three questions that could turn the verified rate into the wrong number: hospital-owned sites claim 57unverified, non-Medicare payment claim 58unverified, and whether a suppressed result can be billed claim 63unverified. It diverges from the desirability deck on sequence. It calls its own condition "the only load-bearing condition across the three decks that can be taken most of the way before a strip exists", because what remains is an internal bill-of-materials estimate. It also notes that the candidate's Verified Composite and Gate Check pre-date the tag moves, and it declines to quote them. Cited fresh: claim 36unconfirmed.
"would the fully-loaded cost of producing one reported potassium-and-creatinine result in an outpatient clinic sit below what the clinic is paid for it" — the load-bearing condition of the viability deck (built 2026-09-29).
Feasibility. Six verified claims anchor this deck, the most of any lens. All six are regulatory-database facts, and it says so: "None speaks to whether the potassium channel works." It leaves two thresholds visibly blank because no claim sets them. One is the allowable potassium error claim 39unconfirmed; the other is the usable window between detection sensitivity and invalid rate. It frames the two as pulling against each other: a lower detection threshold protects accuracy and raises the invalid rate. It names a human-read capillary-versus-venous literature search as the cheapest first look claim 67unverified. Its condition is the point where the other two decks' conditions land. Cited fresh: claim 23unconfirmed, claim 40unconfirmed.
"what share of fingerstick attempts, by ordinary clinic staff on real HF and CKD patients, end with the potassium result suppressed rather than reported, at a detection threshold that still keeps reported results accurate enough to titrate on" — the load-bearing condition of the feasibility deck (built 2026-09-29).
At a hemolysis-detection threshold strict enough that the potassium results the strip does report are accurate enough to titrate on, what share of fingerstick attempts by ordinary outpatient clinic staff on heart-failure and CKD patients end suppressed? Is that share low enough that a prescriber still acts within the visit, and that each reported result still costs the clinic less than the $4.76 plus $5.12 it is paid claim 23unconfirmed, claim 28unconfirmed, claim 36unconfirmed, claim 14verified?
Naming it is not a recommendation, a gate, or a kill. It is the answer to "if a human could only read one more thing about this candidate, what question should it answer." It is drawn from the feasibility deck's condition because the other two lenses' conditions both multiply against it.
_shared/plain-language.md